View Full Version : PowerMac DDR Clarification
arn
Aug 14, 2002, 11:50 AM
xlr8yourmac.com provides clarification on the hot-topic of DDR in the new PowerMacs:
I'm getting lots of mails on this but as you can clearly see from the Apple PowerMac G4 specs page - they note the system bus bandwidth at 1.3GB/sec - clearly indicating that is *not* a DDR system/CPU (frontside) bus. Just like the Xserve, the new G4s use a single data rate system/cpu bus and a DDR memory bus.
3rdpath
Aug 14, 2002, 12:14 PM
so is railhead design...a simple explanation of the whole shebang...
http://www.railheaddesign.com/
Arcady
Aug 14, 2002, 12:30 PM
What I want to know is this: Does the system controller have one CPU bus at 167MHz that is shared by both G4's, or are there two 167MHz busses, each talking to a G4 at 167MHz? If there are two busses for the CPU's, then it would improve things a bit, since both CPU's would have fullspeed access to the memory at their bus speed.
Does the Xserve have two CPU busses, or just one shared bus? Anyone know the answers to these questions? The guesswork stuff isn't helpful.
Thanks :D
porovaara
Aug 14, 2002, 12:33 PM
Originally posted by Arcady
Does the Xserve have two CPU busses, or just one shared bus? Anyone know the answers to these questions? The guesswork stuff isn't helpful.
Thanks :D [/B]
Single.
tortus
Aug 14, 2002, 12:35 PM
Apple threw the DDR onto exisitng motherboards with little changes here and there in an attempt to quell our disatisfaction with the current state of PowerMacs. Sure, the Front Side Bus will be running at 167 mhz, but the CPU is still communicating with the controller at a lower speed than the DDR. All the while, there are PCs out there pushing 500 MHZ FSB. As you can see from the XServe Benchmarks, not much difference between the dual Gigs (sans shiny face plate) and the XServe in terms of performance. By the way, Via is introducing the KT400 chipset that will allow for 200 mhz memory clock speeds. With DDR, we are looking at a 400 mhz memory bus. Apple where are you?
Expect a Badass, new system architecture in spring 2003. Keep the faith brothers and sisters.
topicolo
Aug 14, 2002, 12:41 PM
Hahahaha... Take that cyberfunk! :)
jadam
Aug 14, 2002, 12:41 PM
we knew that
tortus
Aug 14, 2002, 12:46 PM
Keep the funk real.
Peace!
jadam
Aug 14, 2002, 12:48 PM
Originally posted by tortus
Apple threw the DDR onto exisitng motherboards with little changes here and there in an attempt to quell our disatisfaction with the current state of PowerMacs. Sure, the Front Side Bus will be running at 167 mhz, but the CPU is still communicating with the controller at a lower speed than the DDR. All the while, there are PCs out there pushing 500 MHZ FSB. As you can see from the XServe Benchmarks, not much difference between the dual Gigs (sans shiny face plate) and the XServe in terms of performance. By the way, Via is introducing the KT400 chipset that will allow for 200 mhz memory clock speeds. With DDR, we are looking at a 400 mhz memory bus. Apple where are you?
Expect a Badass, new system architecture in spring 2003. Keep the faith brothers and sisters.
GRRR but the athlon can only use a 133mhz bus, the KT400 adds nooo benefit AT ALL with current Athlons!!!! look at the Nforce bencmarks, i mean seriously nforce is dual channel hence 128bit and just barely does better than single channel 266mhz motherboards.
Mosco
Aug 14, 2002, 12:52 PM
Apples explanation taken from maccentral:
"What's different in our architecture from the PC architecture -- which has a higher marketing spec number for the bus -- is that they have every single thing on the system competing for that bus," said Greg Joswiak, senior director of hardware product marketing at Apple. "It has to move quickly because it has a lot of traffic and congestion on the bus. We have each part of the system with its own dedicated bus to the system controller. That means these things don't have a latency or congestion as they wait for other data to migrate through the bus."
tortus
Aug 14, 2002, 12:57 PM
The new Athlons will be sporting a 166 FSB with rumors of 200 fsb not too far off. The point I was trying to make concerned Apple's progress and not the FSB of the AMD systems. Apple comes out with a Memory bus at 333 and other companies are simultaneously 1-upping Apple. The systems are still wonderful and great, but they are a step behind in terms of the latest and the greatest. Hopefully, they take a big leap ahead next year.
ImAlwaysRight
Aug 14, 2002, 01:01 PM
The railheaddesign.com explanation (link above) is the best I have seen.
Why do people keep bringing up XServe tests? The XServe uses ATA/66 and a 133MHz FSB. Folks, the new mid- and high-end PowerMacs have ATA/100 and 167MHz FSB, plus have faster DDR at 333MHz. This will be faster than the XServe taken the combination of these three upgrades, not to mention the better video card. Yes, the new dual gig will be faster than the old dual gig, plus it costs $500 less. Overall, it is a good speed improvement (not fantastic, but good).
deepkid
Aug 14, 2002, 01:04 PM
aromac
macrumors newbie
GOT MY NEW POWERMAC!!! DUAL 1GHZ
So far everything running incredibly smooth. I ran my logic audio on it and it runs it effortlessly. Damn thing is a noise monster though. The fan kicked in out of nowhere for a few seconds and scared the **** out of me.
===
Would you mind taking a look inside at your CPU to whether or not if it says which Motorola chips were used? I'm not sure if it would specify 7455 or 7470...but would be cool if this could be cleared up (since you already own a new powermac). Would system profiler give this detail, as well?
Thanks much!
firewire2001
Aug 14, 2002, 01:09 PM
Originally posted by deepkid
aromac
macrumors newbie
GOT MY NEW POWERMAC!!! DUAL 1GHZ
So far everything running incredibly smooth. I ran my logic audio on it and it runs it effortlessly. Damn thing is a noise monster though. The fan kicked in out of nowhere for a few seconds and scared the **** out of me.
===
Would you mind taking a look inside at your CPU to whether or not if it says which Motorola chips were used? I'm not sure if it would specify 7455 or 7470...but would be cool if this could be cleared up (since you already own a new powermac). Would system profiler give this detail, as well?
Thanks much!
::sigh:: i remember being a newbie.. sorta..
hey, you prolly meant to use the quote function.. you click on the quote button up above and type what you want, or if yer using ie you type [quote!] [\quote!] (without the "!"'s)..
who the heck posted that quote anyways.. "So far everything running incredibly smooth. I ran my logic audio on it and it runs it effortlessly. Damn thing is a noise monster though. The fan kicked in out of nowhere for a few seconds and scared the **** out of me."?
deepkid
Aug 14, 2002, 01:15 PM
Originally posted by firewire2001
::sigh:: i remember being a newbie.. sorta..
hey, you prolly meant to use the quote function.. you click on the quote button up above and type what you want, or if yer using ie you type [quote!] [\quote!] (without the "!"'s)..
who the heck posted that quote anyways.. "So far everything running incredibly smooth. I ran my logic audio on it and it runs it effortlessly. Damn thing is a noise monster though. The fan kicked in out of nowhere for a few seconds and scared the **** out of me."?
aaaaAAAh!! Well I belong to too many different boards and code too much crap each day! :) Thanks for the tip, however.
Actually I snatched that from the incredibly long Powermac introduction thread..figured it would be easier for Aromac to see his post replied to in this new thread.
Some people were saying that the chips are 7470 (one guy said he friend's father is a higher-up at Mot in TX and verified, etc)..just wanted someone with a machine to take a peek.
Been a pretty interesting discussion, thus far.
Sun Baked
Aug 14, 2002, 01:15 PM
This is interesting.
Scroll down to the picture and read the post.
http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=5110932035
jadam
Aug 14, 2002, 01:18 PM
err that was cyberfunk, i dont thin kwe can trust cyberfunk
jadam
Aug 14, 2002, 01:24 PM
Originally posted by Sun Baked
This is interesting.
Scroll down to the picture and read the post.
http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=5110932035
HRRRMMRMRMRMMMMM
Originally posted by jadam
errrr.... one thing. The low end powermac is a sweet deal.
Now why have DDR if it doesnt increase bandwith to the FSB? well lets see. If you were using an old Powermac and lets say you were using the video card and 2 PCI cards and FireWire and USB and such, now you have to u se some of your bandwith for those proccess dont you? sure you do, and guess what, the 2 processors dont get 1.1gb/sec of bandwith but instead get much less.
now with DDR you have more memory to go around, you have a dedicated 1.3gb/sec for the processors and another 1.3gb/sec for say the hard drives and the AGP video card and the PCI cards and Gigabit Ethernet and FireWire and USB and such.
there are reasons for things :p
anyways, the low end is sweeeeeeeeeeeeeeeeeeeeeeeeeeetttttttt, just wish i could afford it :(
http://forums.macrumors.com/showthread.php?s=&threadid=9405&perpage=25&pagenumber=8
WHAT DID I SAY?????????
no but NO ONE LISTENS TO ME! only when its on Arstechnica do they listen to someone else....... LISTEN TOOOO MEEEE
DavidRavenMoon
Aug 14, 2002, 01:25 PM
Originally posted by tortus
Sure, the Front Side Bus will be running at 167 mhz, but the CPU is still communicating with the controller at a lower speed than the DDR. All the while, there are PCs out there pushing 500 MHZ FSB.
This is a common misconception. Actually the RAM is also being clocked at 167 MHz...DDR RAM is not faster than SDR RAM! However DDR offers twice the memory bandwidth SDR RAM does at the same clock speed.
DDR effectively doubles the bandwidth available by sending data on the falling edge of the clock cycle as well as on the rising edge.
What Apple didn't do was to have two busses, one for each CPU. One explanation I have heard is the G4's need to be on the same bus for SMP.
barkmonster
Aug 14, 2002, 01:25 PM
The XServe uses ATA/100 and a 133MHz FSB.
The thing puzzles me is the explaination of RDRAM (or at least the 400Mhz bus) on the intel site :
P4 1.4 - 2.0Ghz info (http://intel.com/design/pentium4/datashts/249198.htm)
The 400 MHz system bus is a quad-pumped bus running off a 100 MHz system clock making 3.2 GB/sec data transfer rates possible.
That's the old version of the P4 with the 256K L2, the new one's running off the same 133Mhz system clock as the entry level G4 but they've quad pumped it to 533Mhz and doubled the L2 cache size.
I think seeing as we've got 2.7Gb/s of bandwidth to share with our PCI cards, AGP graphics card, ethernet, USB, firewire and airport, not to mention the hard drive controllers are both independent of cpu. We should see some improvement over the old motherboard purely because the system controller seperates the cpu's bandwidth from the components.
I'd love to see how these macs perform in Protools LE!!!
All that bandwidth for the hard drives and pci slots has got to have quite an impact on performance.
cyberfunk
Aug 14, 2002, 01:26 PM
Originally posted by topicolo
Hahahaha... Take that cyberfunk! :)
Now now...
jadam
Aug 14, 2002, 01:28 PM
Originally posted by DavidRavenMoon
This is a common misconception. Actually the RAM is also being clocked at 167 MHz...DDR RAM is not faster than SDR RAM! However DDR offers twice the memory bandwidth SDR RAM does at the same clock speed.
DDR effectively doubles the bandwidth available by sending data on the falling edge of the clock cycle as well as on the rising edge.
What Apple didn't do was to have two busses, one for each CPU. One explanation I have heard is the G4's need to be on the same bus for SMP.
because they need to be sycnronized... err
deepkid
Aug 14, 2002, 01:31 PM
Originally posted by Sun Baked
This is interesting.
Scroll down to the picture and read the post.
http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=5110932035
The Natebrau system architecture explanations are pretty good applesauce for us novices! I had to reread it a few times, but it does make sense.
It does make me wonder what hurdles Apple'd have to clear in order to have an independent bus for each chip (7470s required?), how much improvement we'd see... and how much that'd cost, etc. etc.
Thanks for the reference.
cyberfunk
Aug 14, 2002, 01:31 PM
Originally posted by jadam
err that was cyberfunk, i dont thin kwe can trust cyberfunk
What was me???
cyberfunk
Aug 14, 2002, 01:34 PM
What have I said thats untrustful ?
deepkid
Aug 14, 2002, 01:35 PM
Originally posted by DavidRavenMoon
What Apple didn't do was to have two busses, one for each CPU. One explanation I have heard is the G4's need to be on the same bus for SMP.
Interesting... does this possibly mean that multiple CPUs would always have to be on the same bus, going foward? Or could the os be amended in such a ways to support 1 bus/CPU?
Thanks.
topicolo
Aug 14, 2002, 01:42 PM
Originally posted by jadam
HRRRMMRMRMRMMMMM
http://forums.macrumors.com/showthread.php?s=&threadid=9405&perpage=25&pagenumber=8
WHAT DID I SAY?????????
no but NO ONE LISTENS TO ME! only when its on Arstechnica do they listen to someone else....... LISTEN TOOOO MEEEE
I know what you mean :D
jadam
Aug 14, 2002, 01:43 PM
ARRRR read my posts .............
no the G4 processor would not allow apple to use 2 different busses for each processor. I mean, they COULD, but... that would be horribley inefficient and would just be stupid. IT would be like a BeoWolf Cluster in a case.
topicolo
Aug 14, 2002, 01:45 PM
Originally posted by cyberfunk
What have I said thats untrustful ?
It was your many posts saying that the powermac was fully DDR while I kept saying the opposite. I hate to say I told you so...
I told you so! :D
jadam
Aug 14, 2002, 01:47 PM
Originally posted by topicolo
It was your many posts saying that the powermac was fully DDR while I kept saying the opposite. I hate to say I told you so...
I told you so! :D
hahahaha
Sun Baked
Aug 14, 2002, 01:52 PM
Originally posted by jadam
WHAT DID I SAY?????????
no but NO ONE LISTENS TO ME! only when its on Arstechnica do they listen to someone else....... LISTEN TOOOO MEEEE
I read BOTH posts, I though they basically said the same thing. But natebrau's post included the pretty picture. :)
His post also happened to be easiest one to find, and a darn good explanation.
cyberfunk
Aug 14, 2002, 02:42 PM
/Harumph, you cant blame someone for being too optimistic, can ya ?>
Specially seeing as I just ordered one, I just think your all jealous.
jadam
Aug 14, 2002, 02:49 PM
Originally posted by cyberfunk
/Harumph, you cant blame someone for being too optimistic, can ya ?>
Specially seeing as I just ordered one, I just think your all jealous.
haha which one?
anyhoo i got a new ibook and i love it :p
topicolo
Aug 14, 2002, 03:07 PM
Oh well, I'm not all that into the desktop powermacs anyway. I don't really need speed as much as portability. I'm gonna get another Powerbook! (my duo 2300 is only good for word processing and playing gameboy games :)) (besides, my pc already has 512mb PC2700DDR on a true 133DDR bus (overclocked to 166Mhz /333DDR effective)
sford
Aug 14, 2002, 03:36 PM
I'm not trying to start something but...
No one seems all that happy the the improvement made by Apple, just how fast is fast enough? I assumed (ya i know..ass..) that most Apple users were better (smarter) than PC weenies. I'm currious as to what can't you realistically accomplish in a reasonable time with any of the new macs. ( except Photoshop )
I just wish I could afford one.....
Regards,
topicolo
Aug 14, 2002, 03:38 PM
Play Quake 3 at 1600x1200 with 4x FSAA at a reasonable frame rate :-)
locovaca
Aug 14, 2002, 03:42 PM
Originally posted by barkmonster
The XServe uses ATA/100 and a 133MHz FSB.
The thing puzzles me is the explaination of RDRAM (or at least the 400Mhz bus) on the intel site :
P4 1.4 - 2.0Ghz info (http://intel.com/design/pentium4/datashts/249198.htm)
That's the old version of the P4 with the 256K L2, the new one's running off the same 133Mhz system clock as the entry level G4 but they've quad pumped it to 533Mhz and doubled the L2 cache size.
I think seeing as we've got 2.7Gb/s of bandwidth to share with our PCI cards, AGP graphics card, ethernet, USB, firewire and airport, not to mention the hard drive controllers are both independent of cpu. We should see some improvement over the old motherboard purely because the system controller seperates the cpu's bandwidth from the components.
I'd love to see how these macs perform in Protools LE!!!
All that bandwidth for the hard drives and pci slots has got to have quite an impact on performance.
The difference between Intel, AMD, and Apple is this:
Intel: 100 mhz QDR (transfers data 4 times per clock cycle, making it an effective 400 mhz clock)
AMD: 133 mhz DDR (transfers data 2 times per clock cycle, making it an effective 266 mhz clock)
Apple: 166 mhz SDR (only transfers data once per clock cycle)
Without an architecture change, you can't make a processor that uses a SDR bus go to a DDR bus. You could make it a 333 mhz SDR bus, but it's highly doubtful that the processor would be able to handle that.
Now, the new P4's do use a 133 QDR bus, however, their own chipset still only has a memory bus of 100 mhz QDR, which does still help, but not as much as it could. Remember, even though you have 1.3 GB/s to give to your I/O devices, every I/O device still uses the CPU quite a bit, and generates a lot of traffic on that FSB. Additionally SMP systems generate a lot of traffic on the FSB communicating to each other, and so you end up with only 900 megs of bandwidth left over.
There is no such thing as an I/O device that is independent of a CPU. It will use the CPU in some way, at the bare minimum raising an interrupt everytime data is transferred. You can offload some of the processing, but it's all but impossible to offload everything.
Rocketman
Aug 14, 2002, 04:13 PM
Apple seems to feel that since there are separate busses for memory, PCI, AGP, ATA, and main bus that the main bus can be 133-166 and the computer is still double the practical speed as a wintel P4 PC.
This may be true indeed.
However why not upgrade the syatem bus substantially anyway? I cannot believe it is not practical since 266-400 mhz system busses have neen on wintel boxes for some time now, and even if you assume Apple's adoption should be one year behind the curve it should at least have 266-300 mhz.
I for one hope this lack of emphasis on system bus which IS a major point of customer concern and complaint, is because they have some form of hypertransport on the horizon.
The current system can have 4 ATA drives (2x100, 2x66) plus two optical drives but optimum performance is achieved if you only have one drive on each bus, one on a 66 and one on a 100 ATA bus.
So there are limitations.
On the positive side Apple has as expected delivered a SUBSTANTIAL improvement in overall speed for NO INCREASE in pricing.
Say what you will about what Apple "shoulda done", but it is leading the industry in technology, speed, stability, new application development and profitability. You gotta respect that.
Are you a cheapskate? Buy an iBook G3 or a CRT iMac G3. $1100 and $749 respectively.
Rocketman
Sun Baked
Aug 14, 2002, 04:15 PM
Originally posted by locovaca
The difference between Intel, AMD, and Apple is this:
Intel: 100 mhz QDR (transfers data 4 times per clock cycle, making it an effective 400 mhz clock)
AMD: 133 mhz DDR (transfers data 2 times per clock cycle, making it an effective 266 mhz clock)
Apple: 166 mhz SDR (only transfers data once per clock cycle)
Could you PLEASE include the width of each companies data path with your example, otherwise you're just proving which of the three operates at the highest effective frequency.
It would be nice to know which expample you're providing actually MOVES the most data.
---
If you convert your table to this
Truck A goes 65 mph
Truck B goes 50 mph
Truck C goes 35 mph
You're concluding Truck A is the best performer based on speed alone.
But if you a little more data.
Truck A (F-150) goes 65 mph and carries .75 tons
Truck B (UPS truck) goes 50 mph and carries 3 tons
Truck C (a loaded semi) goes 35 mph and carries 24 tons
The actual results may not be tied to speed alone.
----
Fun thing is once you get over realizing it's not MHz alone that wins the race, they'll add another layer of fuzz to the equation. ;)
Hemingray
Aug 14, 2002, 05:35 PM
Well hm, while we're on the subject of "I told you so"'s, I'd just like to deliver the following " :p " to AmbitiousLemon (no hard feelings, of course! :D ) We all smeg up once in awhile.
Originally posted by AmbitiousLemon
no ddr fsb? not sure what you mean by that and i suspect neither do you. apple upped the speed of the fsb. it fully supports ddr.
soilchmst
Aug 14, 2002, 06:12 PM
Originally posted by topicolo
Play Quake 3 at 1600x1200 with 4x FSAA at a reasonable frame rate :-)
since Quake 3 is optimized for dual proccessors and the Geforce Ti is pretty damn fast and can access the the DDR directly, wouln't Q3 run stellarly? ;)
rice_web
Aug 14, 2002, 07:02 PM
The one area that will shine with the new PowerMacs is games, despite not having a full DDR implementation. With the AGP bus now independent of the CPU, framerates will skyrocket.
alex_ant
Aug 14, 2002, 07:44 PM
We know that the new Power Mac has a crippled DDR implementation. Could the 7455 CPU module in one of these systems be replaced in the future with a different CPU that DOES support DDR, thus un-crippling the memory architecture?
In other words, was DDR hacked onto the motherboard in order to achieve a real performance increase, or was it hacked onto the motherboard in order to facilitate future CPUs and their respective modules that would be able to truly exploit it, providing an upgrade path to a certain future DDR-compatible PowerPC chip?
Alex
iH8Quark
Aug 14, 2002, 07:56 PM
Sadly, I'm certain that this means that we won't be seeing anything resembling a "G5" until MWNY 2003. The next rev will be speed bumped processors on this new board. Apple wouldn't eat all that R&D. It makes no sense. That means we're stuck with this BUS bottleneck.
IMHO, I don't think this is what the "pro" users and video effects and compositing people were waiting for.
jrbohorquezg
Aug 14, 2002, 08:02 PM
Guys, I'm sorry to post this message because it's a little off-topic, but I kind of need your expertise:
My problem is: should I get the new DP 1Ghz or the old DP 1Ghz? do you think that there's a real performance difference?
Thanks for your comments!
alex_ant
Aug 14, 2002, 08:10 PM
Originally posted by jrbohorquezg
Guys, I'm sorry to post this message because it's a little off-topic, but I kind of need your expertise:
My problem is: should I get the new DP 1Ghz or the old DP 1Ghz? do you think that there's a real performance difference?
Thanks for your comments!
If your only consideration is performance, then if you could find the old DP 1GHz for substantially less money than the new one, I would go with that. I would be very surprised to see more than a 5% performance difference between new and old. The new system has a hacked DDR implementation the same as the Xserve, with the same CPU (MPC7455) as both the Xserve and the old DP machine. Various performance benchmarks between the old DP Power Mac and the DP Xserve revealed virtually no performance difference, indicating that this crippled DDR implementation does little to nothing to enhance performance.
HOWEVER (and the following is only speculation), the new case of the new Power Macs may be a hint at some sort of possible upgrade to this Power4-derived PPC we're all waiting for. It would make sense - why give all this cooling and ventilation to a G4 system that doesn't need it? Still though, I would buy the old G4, because there's always eBay.
Alex
cyberfunk
Aug 14, 2002, 08:19 PM
Originally posted by topicolo
Play Quake 3 at 1600x1200 with 4x FSAA at a reasonable frame rate :-)
Well.. thats the graphics card mostly..
eirik
Aug 14, 2002, 08:19 PM
Originally posted by jrbohorquezg
Guys, I'm sorry to post this message because it's a little off-topic, but I kind of need your expertise:
My problem is: should I get the new DP 1Ghz or the old DP 1Ghz? do you think that there's a real performance difference?
Thanks for your comments!
Obviously, independent performance tests will indicate to you whether the price differential is worth the performance differential.
There will be a significant performance differential for two main reasons:
First, the data throughput from the main memory to the CPU's will be 325 MB/s faster on the new dual 1 gig than on the old dual 1 gig.
Second, because the AGP bus/connection to the System Controller is independent of the separate bus/connector from the System Controller to the CPU's, your video can speak directly to the main memory and supposedly can speak fluent DDR. That is, you should realize twice the data throughput from data memory to video card than from DDR to the CPU's. That's the simple way to look at it. The video card and CPU's to have to coordinate their actions with one another so that takes away some of the video card b/w. Also, the so-called 'direct' communication from video card to main memory does in some way share b/w with the CPU in sofar as the DDR can crank out data at 2.7 GB/s. If its cranking out 1.3 GB/s to the CPU's then there's that much that it cannot send to the video card or 1.4 GB/s.
On the old dualie, you'd have significantly less data throughput to your video card as well as less to your CPU (both directly and indicerctly in that the bus is shared by both the CPU and video card I believe).
So, with Quartz Extreme, there should be a very significant difference in performance, especially in terms of snappy response.
Conclusion: we should expect at least 20% improved performance for main memory intensive applications for the new 1 gig dualies over that of the old 1 gig dualies.
I'm stretching my understanding of these things here so please be kind oh EE-enlightened folk.
cyberfunk
Aug 14, 2002, 08:26 PM
Originally posted by alex_ant
why give all this cooling and ventilation to a G4 system that doesn't need it....
MY bet on the big heat sink is that they're using a larger, slower fan, so it's more effective to have a larger surface area.
eirik
Aug 14, 2002, 08:39 PM
Originally posted by alex_ant
If your only consideration is performance, then if you could find the old DP 1GHz for substantially less money than the new one, I would go with that. I would be very surprised to see more than a 5% performance difference between new and old. The new system has a hacked DDR implementation the same as the Xserve, with the same CPU (MPC7455) as both the Xserve and the old DP machine. Various performance benchmarks between the old DP Power Mac and the DP Xserve revealed virtually no performance difference, indicating that this crippled DDR implementation does little to nothing to enhance performance.
Alex
Alex, it is my understanding that the CPU's in the Xserve speak to the System Controller at 133 MHz or 1.05GB/s of data throughput. The new 1 Gig dualies speaks at 167 MHz or 1.3 GB/s. So, that alone should yield close to a 25% performance increase for MEMORY INTENSIVE operations.
Also, I don't know (don't recall, too lazy to look right now) if those benchmarks were based on 10.1 or a 10.2 beta. It would make for some difference wouldn't it, given Quartz Extreme would off-load some data throughput between the System Controller and CPU?
I'm not CERTAIN about this but I am pretty confident.
alex_ant
Aug 14, 2002, 08:39 PM
Originally posted by cyberfunk
MY bet on the big heat sink is that they're using a larger, slower fan, so it's more effective to have a larger surface area.
Don't say that! Say it's for a G5, damnit! :)
Maybe if we wish hard enough, we can make THESE Power Macs the G5s! Come on everybody, let's do it. "I love these new Power Mac G5s!" Everybody now. "I love these new Power Mac G5s." Mind over matter, people!
eirik
Aug 14, 2002, 08:50 PM
I'm looking forward to hearing from folk with the new PM's as to whether or not the CPU's are soldered in or not. I presume if they're soldered upgrading the CPU becomes a hell of a lot more complicated for those of us who have never soldered electronic components before, let alone even own the appropriate soldering iron.
Also, if the son of Power4 draws significantly more power than the PPC 7455, won't that require some other adjustment that most consumers are not up to doing? Does anybody have any educated guesses about the pin count/sockets of junior also?
I'm not too bullish on this upgrade speculation, though a rather learned person on arstechnica believes an upgrade to a 7470 or some other 74xx w/ native DDR capability is quite probable. It would still require soldering and over-clocking, however.
Eirik
DavidRavenMoon
Aug 14, 2002, 09:42 PM
Originally posted by jadam
because they need to be sycnronized... err
Is that anything like "synchronized"...err?
;)
Sun Baked
Aug 14, 2002, 10:17 PM
Originally posted by eirik
I'm stretching my understanding of these things here so please be kind oh EE-enlightened folk.
The only thing an EE would tell you is you may be looking at the wrong end of the problem and making it way more confusing.
Everyone is focusing on the FSB (between the controller and the CPU), focus on the datapath between the memory and controller instead.
There is much more memory bandwidth here on the DDR path for the system controller to share via DMA processes. You have the CPU, AGP card, PCI cards, and the primary & secondary ATA buses all fighting for their share of memory's bandwidth.
Even if you give the FSB all the data it wants, there's still enough bandwidth left for the rest of the DMA requests.
Looking at it this way it might not sound so bad, but a slow FSB also limits the maximum number of operations the CPUs can handle - and the PPCs are a lot hungrier per clock cycle than the avereage wafer. And you're trying to feed two of these little beasties.
Rocketman
Aug 14, 2002, 10:54 PM
Originally posted by alex_ant
We know that the new Power Mac has a crippled DDR implementation. Could the 7455 CPU module in one of these systems be replaced in the future with a different CPU that DOES support DDR, thus un-crippling the memory architecture?
In other words, was DDR hacked onto the motherboard in order to achieve a real performance increase, or was it hacked onto the motherboard in order to facilitate future CPUs and their respective modules that would be able to truly exploit it, providing an upgrade path to a certain future DDR-compatible PowerPC chip?
Alex
Apple has officially stated (in confidence) thatthe current box is G5 compatible FWIW.
Jerry
Rocketman
Aug 14, 2002, 10:58 PM
Originally posted by alex_ant
If your only consideration is performance, then if you could find the old DP 1GHz for substantially less money than the new one, I would go with that. I would be very surprised to see more than a 5% performance difference between new and old. The new system has a hacked DDR implementation the same as the Xserve, with the same CPU (MPC7455) as both the Xserve and the old DP machine. Various performance benchmarks between the old DP Power Mac and the DP Xserve revealed virtually no performance difference, indicating that this crippled DDR implementation does little to nothing to enhance performance.
HOWEVER (and the following is only speculation), the new case of the new Power Macs may be a hint at some sort of possible upgrade to this Power4-derived PPC we're all waiting for. It would make sense - why give all this cooling and ventilation to a G4 system that doesn't need it? Still though, I would buy the old G4, because there's always eBay.
Alex
While I find each use of the word "crippled" appropriate, I remind you that the throughput and the Gigaflop rating is about double. That matters.
Alot.
Rocketman
wildcat4100
Aug 15, 2002, 01:20 AM
Originally posted by topicolo
Play Quake 3 at 1600x1200 with 4x FSAA at a reasonable frame rate :-)
Try play WarCraft 3 with 1600x1200 and all the option set to high, let see what the frame rate you're getting
DannyZR2
Aug 15, 2002, 03:50 AM
So we know the previous 1ghz duallies ran up to 15 gigaflops according to apple.. the new 1.25ghz with ddr ram and 167mhz fsb run up to 18.3 gigaflops, so we know there is an improvement..
but I want to know what the new 1ghz duallies with 167 bus and ddr will do... or even still, what the xserve will do, since it's very similar to the new mid powermac (aside the 167mhz)
I'd hope the new dual 1ghz would do at least 15gflops, like 16 or so, but I'd like to see that spec!
DannyZR2
Aug 15, 2002, 03:54 AM
One thing I found interesting and helpful in my purchasing decisiong is what I read on arstechnica...
one guy mentioned that since the mobo's are the same all accross the line, it shouldn't be much to overclock the dual 867 to have a 167mhz bus and the cpu to 900 somthing.. (can't remember the exact multiplier) It sounds feasible, espescially considering the ibooks that can be overclocked via software from 600 to 700mhz!!!
Aren't there utilities that will over clock bus speed??
topicolo
Aug 15, 2002, 09:35 AM
Originally posted by soilchmst
since Quake 3 is optimized for dual proccessors and the Geforce Ti is pretty damn fast and can access the the DDR directly, wouln't Q3 run stellarly? ;)
not nearly good enough to make 60fps, even on a 2.54Ghz P4. The 4xFSAA actually downsamples the images from a resolution 4x greater.
Hemingray
Aug 15, 2002, 02:48 PM
Hey guys, if you want some disturbing info, check out BareFeat's new benchmark test comparing the old dual 1GHz with the new DDR dual 1GHz!
http://www.barefeats.com/pmddr.html
To my surprise and chagrin, the new DDR Power Mac has no apparent performance advantage over the old SDRAM Power Mac running at the same clock speed.
Ouch! :eek:
Sun Baked
Aug 15, 2002, 03:12 PM
It's so nice to see people taking a typical Mac OS 9 style test and saying, see the new machine is NOT faster.
Too bad these tests are not real world for OS X.
With the multithreading we all know a lot of these single actions are slower, but the machine is also doing MORE at the same time.
Where are the real OS X tests?
Where are the true tests of the multi-tasking capabilities? Such as Photoshop&Quake Frame Rate, Photoshop&Storage Testing, etc.
Nobody is really showing the true capabilities of the machines, nor the new buses.
Here let me pat you on the head for blindly swallowing this single action OS 9-style line of BS. (PAT, PAT, SLAP)
eirik
Aug 15, 2002, 03:28 PM
Originally posted by Sun Baked
It's so nice to see people taking a typical Mac OS 9 style test and saying, see the new machine is NOT faster.
Too bad these tests are not real world for OS X.
With the multithreading we all know a lot of these single actions are slower, but the machine is also doing MORE at the same time.
Where are the real OS X tests?
Where are the true tests of the multi-tasking capabilities? Such as Photoshop&Quake Frame Rate, Photoshop&Storage Testing, etc.
Nobody is really showing the true capabilities of the machines, nor the new buses.
Here let me pat you on the head for blindly swallowing this single action OS 9-style line of BS. (PAT, PAT, SLAP)
Would you mind elaborating on how the Barefeat tests were not realistic and were OS 9 centric? I don't understand.
I looked at the test details and it implied that the Photoshop test was version 7.0 so that it would take advantage of 7.0. I don't know about the Bryce and mp3 (iTunes) tests, however.
Seriously, I don't understand and your post sounds very provocative, like something that I'd like to know too.
Thanks,
Eirik
BS. I emailed Rob over at Barefeats, asking him to run tests with larger files as well as video tests so that we'd see results of tests that were far more memory intensive than the 30 MB PS 7.0 test.
Sun Baked
Aug 15, 2002, 03:55 PM
Eirik,
Under Mac OS 9 you might do each of his tests, listed under "How I test" as a single action - only due to the fact that some of these actions would suck up so much CPU time that doing several at the same time would make GUI rather slow to respond, and you made allowances for this.
Because some of the programs were such hogs you might do something like start a rendering job and go get a cup of coffee. Plus under OS 9 there were a lot of applications that plain would never use the second processor - it just sat there with nothing to do.
But under OS X, the entire system and the applications are designed for SMP and multithreaded to the point where doing multiple tasks at once is becoming the norm.
So basically under OS X, it's a better indication of how the machine/OS works if you do something that stresses the CPU/memory/video/storage in a combination that would test 2, 3, or 4 of these at the same time.
I'm adding this link, because these guys can say it a lot better than I, why these OS 9 style test are pointless on a OS X system.
http://arstechnica.infopop.net/OpenTopic/page?a=tpc&s=50009562&f=8300945231&m=9600944035
Dragoth really lets him have it. :)
deepkid
Aug 15, 2002, 05:23 PM
Originally posted by Hemingray
Hey guys, if you want some disturbing info, check out BareFeat's new benchmark test comparing the old dual 1GHz with the new DDR dual 1GHz!
http://www.barefeats.com/pmddr.html
Ouch! :eek:
Did I miss something? I don't see where it says which OS he used..and if he used OS X, did he use Jaguar? That would see most appropriate, since it is the newest and most revelant OS/version.
rice_web
Aug 15, 2002, 05:35 PM
He didn't even specify the amount of RAM used.
Arcady
Aug 16, 2002, 02:59 AM
Here is a block diagram of the new PowerMacs (from the service manual):
Sun Baked
Aug 16, 2002, 05:19 AM
Thank you Arcady
MacBandit
Aug 16, 2002, 10:15 AM
Originally posted by Rocketman
Apple has officially stated (in confidence) thatthe current box is G5 compatible FWIW.
Jerry
Where has apple stated that this machine is G5 compatible? The only thing I have seen saying anything like that was that PDF someone made up describing the new tower features. The only way you could see the G5 upgradeable in that PDF was to move an image that was overlying it revealing the alternative header.
My question is, does anyone know where the FSB resides within the machine? Is it on the CPU card or is it on the motherboard? If it's on the motherboard I don't see how all the G4 7470's and G5's in the world would give you true DDR. Seeing how the bottleneck and lack of true DDR support is not only in the current CPU's but because of the CPU the FSB is also limited. So you have to not only replace the CPU but also the FSB.
I don't know about anyone else but I've been running a B/W G3 400 for too long now and have already ordered the new Dual 1gig it doesn't matter if it's not a G5 it's a revolutionary jump for anyone still running a G3 and some that are running G4's.
I personally really like the new case and I want to settle a few disputes that people are having but have been answered many times but they haven't heard.
1. People keep saying this isn't truly the prototype case we all saw because the prototype was white.
A) The prototype was not white but still silver the people who took the pictures of the prototype even posted it was silver and apologized for the picture is had to do with overexposure of the image.
2. The new vents and larger heat sink are not for a new cpu. They are simply there to make the new machine much quieter in normal use. People you've all been complaining about how loud the Quicksilver is and now that Apple has addressed the problem you can't even recognise it all you say is how ugly it is. This machine is designed to move much more air without making any noise in normal use. Yes when the VARIABLE FAN spins up to full speed it does make a quite a bit of noise but not really any more then the Quicksilvers.
Apple has provided us with a completely new machine whether you like it or not. It is not a lot faster for normal tasks but when you start editing large data files like video etc. it should be significantly faster. Also this computer has added a lot of functionality not only is it quieter but it also has an ATA 100 bus and an ATA 66 bus for support of 4 high speed drives without needing to fill anymore PCI slots.
I repeat this is a great leap forward for the proffesionals for who it is made for. So if you can't see this quit whinning and get an iMac or that friggin PC you keep talking about. Just remember that PC does not run OSX.
alex_ant
Aug 16, 2002, 10:47 AM
Originally posted by MacBandit
2. The new vents and larger heat sink are not for a new cpu. They are simply there to make the new machine much quieter in normal use. People you've all been complaining about how loud the Quicksilver is and now that Apple has addressed the problem you can't even recognise it all you say is how ugly it is. This machine is designed to move much more air without making any noise in normal use. Yes when the VARIABLE FAN spins up to full speed it does make a quite a bit of noise but not really any more then the Quicksilvers.
Actually, everyone I've heard who has used the new machines so far says that they're even louder than the Quicksilvers. "Very loud" as some put it. That surprised me, because I too thought the new machine would be quieter.
Alex
vBulletin® v3.8.6, Copyright ©2000-2012, Jelsoft Enterprises Ltd.